Business process analysis device, business process analysis method, and business process analysis program

Abstract

To enable proper analysis of a creation activity time in tasks that make up a business process.SOLUTION: A business process analysis device 10 for analyzing a business process based on deliverable data created by an application includes a processor 11. The processor 11 is configured to obtain a creation date and time when the deliverable data is created, obtain an application log including an active time during which the deliverable data is active in the application, and estimate a creation activity time, which is the time during which an activity for creating a deliverable in a task of the business process for creating the deliverable data is performed, based on the creation date and time of the deliverable data and the active time.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a technique for analyzing processes (also called business processes) carried out in various types of business. [Background technology]

[0002] Digital transformation (DX), as typified by the introduction of robotic process automation (RPA), has been attracting attention as an effort to improve business efficiency and productivity. RPA is software that uses artificial intelligence (AI) and machine learning to perform tasks that have previously only been handled by humans, or more advanced tasks, on behalf of humans. DX refers to companies using digital technology to transform their organizations and business models. In these DX efforts, business processes have been visualized in order to identify inefficient tasks that need to be improved or transformed (see Patent Document 1, for example).

[0003] However, to identify these business processes, a specialist with business analysis skills must observe the organization where the business is carried out, interview the business personnel, read and understand the business manuals and standards, transcribe the contents manually, and then check and revise the transcribed contents with various people involved in the business, which requires a lot of time and effort. Also, depending on the level of understanding of the business of the interviewees, exceptional business operations that are performed infrequently may be overlooked. Furthermore, the more advanced the business, the more rapidly the way the business is carried out continues to change. For this reason, it is not enough to simply identify the business process at a certain point in time, but it is necessary to keep up with the changes.

[0004] Therefore, in recent years, process mining has been attracting attention as a technology for automating the identification of business processes. This is a technology / method for estimating tasks performed in a business and the order of their execution (business process) from logs (operation history = event data) in IT systems used to proceed with the business, as shown in Patent Document 2 and Non-Patent Document 1, for example. A task is one element that constitutes a business process. For example, in a procurement business process, there are tasks such as "create an order specification" and "check inventory," and in a design business process, there are tasks such as "design the shape of a XX part," "evaluate the strength of a XX part," and "create materials for design review." This process mining technology can automate or reduce the labor required for business analysis that previously required experts to perform, and can capture changes in business processes by continuing to use it. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2003-150650 A [Patent Document 2] JP 2022-165851 A [Patent Document 3] JP 2011-164936 A [Non-patent literature]

[0006] [Non-Patent Document 1] Tadashi Iijima, Keiichi Tabata, Shinobu Saito, Process Mining Survey (1st: Overview and Basic Concepts), Journal of Information Systems Society, 2016. [Non-Patent Document 2] Nachan, NS, Nachan, SS, Microsoft Viva for Everyone. In: Up and Running on Microsoft Viva Connections., Apress, Berkeley, CA. https: / / doi.org / 10.1007 / 978-1-4842-8606-7_1. Summary of the Invention [Problem to be solved by the invention]

[0007] However, the above techniques have the following problems.

[0008] In Patent Document 1, the identification of business processes needs to be performed by an expert. In the technology described in Non-Patent Document 1, it is necessary for an IT (Information Technology) system to be able to generate and acquire event logs corresponding to all tasks that make up a business process.

[0009] The technology described in Patent Document 2 discloses a means for estimating tasks based on the similarity of deliverable document files and estimating the execution order of the tasks. However, it is not possible to estimate the time spent performing a task (task activity time). In order to calculate the task activity time, the start and end times of the task must be recorded, but in file systems such as Windows (registered trademark) and OneDrive, only the save date and time, i.e., the end date and time, are recorded, and the task activity time cannot be obtained.

[0010] The technology described in Non-Patent Document 2 can calculate the total operating time of an application (such as Excel or Outlook) and make recommendations for each individual's activities, but it cannot estimate task activity time in a business process because it does not know what task the same Excel was used for, such as cost calculation or intensity calculation.

[0011] The technology described in Patent Document 3 discloses a means for acquiring a PC (Personal Computer) operation log, recognizing a creation activity if there is a keyboard operation during the period when a file is open, and a reference activity if there is no keyboard operation, and recording the time when the file was active as the creation activity time and the reference activity time. However, Patent Document 3 does not disclose how to estimate which file was referenced when which file was created. In addition, even if the same or similar file is created, the task differs depending on who created the file. For example, in a business flow in which the task of "creating a design document" and the task of "reviewing the design document" are performed in sequence, the person in charge has the role of creating the design document and the superior has the role of reviewing the design document. In this case, since it is a creation activity for the same file, it cannot be separated into the task of "creating a design document" and the task of "reviewing the design document". In addition, since the creation activity or the reference activity is determined based on the keyboard operation log, for example, if a keyboard input is performed for searching during a reference activity, it is recognized as a creation activity even if only a reference is performed, so the accuracy of determining whether it is a creation activity or a reference activity is insufficient. Furthermore, since it is necessary to capture keyboard operations, it is unavoidable to record sensitive information such as passwords. From a security standpoint, however, it is desirable to avoid capturing keyboard operation logs.

[0012] The present invention has been made in consideration of the above circumstances, and has an object to provide a technique that enables appropriate analysis of production activity time in tasks that constitute a business process. [Means for solving the problem]

[0013] In order to achieve the above-mentioned object, a business process analysis device according to one aspect is a business process analysis device that analyzes a business process based on deliverable data created by an application, and is equipped with a processor, wherein the processor acquires a creation date and time when the deliverable data was created, acquires an application log including an active time during which the deliverable data is active in an application, and estimates a creation activity time, which is the time during which an activity to create a deliverable in a task of the business process that creates the deliverable data, was carried out based on the creation date and time of the deliverable data and the active time. Effect of the Invention

[0014] It is possible to properly analyze the creation activity time for the tasks that make up a business process. [Brief description of the drawings]

[0015] [Figure 1] FIG. 1 is a hardware configuration diagram of a business process analysis system according to an embodiment. [Diagram 2] FIG. 2 is a functional configuration diagram of a business process analysis system according to an embodiment. [Diagram 3] FIG. 3 is a diagram illustrating an example of a deliverable database according to an embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of event log data according to an embodiment. [Diagram 5] FIG. 5 is a diagram illustrating a process of estimating a business process according to an embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of business process data according to an embodiment. [Figure 7] FIG. 7 is a diagram illustrating an example of an application log according to an embodiment. [Figure 8] FIG. 8 is a PAD (Problem Analysis Diagram) diagram showing a task activity analysis process according to one embodiment. [Figure 9] FIG. 9 is a PAD diagram illustrating a task separation process according to one embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of business process data after the task separation process according to an embodiment. [Figure 11] FIG. 11 is a diagram showing an example of a business process analysis result screen according to an embodiment. [Figure 12] FIG. 12 is a flowchart of the overall processing of a business process analysis device according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] The following embodiments will be described with reference to the drawings. Note that the following embodiments do not limit the scope of the invention, and not all of the elements and combinations thereof described in the embodiments are necessarily essential to the solution of the invention.

[0017] FIG. 1 is a hardware configuration diagram of a business process analysis system according to an embodiment.

[0018] The business process analysis system 1 includes a business process analysis device 10, one or more business PCs 20 (20-1, 20-2, ... 20-n, etc.) connected to the business process analysis device 10 via a network 2, and an output device 18 and an input device 19 connected to the business process analysis device 10.

[0019] The network 2 may be a Wide Area Network (WAN) such as the Internet, a wireless or wired Local Area Network (LAN), or a combination of a WAN and a LAN.

[0020] The work PC 20 is an example of an information processing device, and is configured, for example, by a computer such as a PC (Personal Computer) having a processor, memory, etc. The work PC 20 is a device used by a user, and is a device on which applications related to a business process are executed. In the work PC 20, the user uses the applications to create deliverable data for tasks of the business process, and stores the created deliverable data in a deliverable database 101, which will be described later.

[0021] The output device 18 is, for example, a display device such as a liquid crystal monitor or an organic EL (Electro Luminescence) display, an audio output device such as a speaker, a printing device such as a printer, etc. The input device 19 is, for example, a keyboard, a mouse, a touch panel, a card reader, or an audio input device, etc.

[0022] The business process analysis device 10 estimates and analyzes business processes based on deliverable data. The business process analysis device 10 is configured with a computer such as a PC or a general-purpose server, and has a processor 11, a communication control device 12, a communication interface 13 (indicated as I / F in FIG. 1), a main memory device 14, an auxiliary memory device 15, an internal bus 16, and an input / output interface 17 (indicated as I / O in FIG. 1). The processor 11, the communication control device 12, the communication interface 13, the main memory device 14, the auxiliary memory device 15, and the input / output interface 17 are connected to each other via the internal bus 16 so as to be able to communicate with each other.

[0023] The processor 11 is hardware that controls the overall operation of the business process analysis device 10. The processor 11 is not limited to a CPU (Central Processing Unit), but may be a GPU (Graphics Processing Unit) or an ASIC (Application Specific Integrated Circuit), or may include these. The processor 11 reads out programs stored in the auxiliary storage device 15 into the main storage device 14 and executes them to perform various processes.

[0024] The primary storage device 14 is, for example, a semiconductor memory such as a static random access memory (SRAM) or a dynamic random access memory (DRAM). The primary storage device 14 stores programs executed by the processor 11 and various data used by the processor 11 for processing.

[0025] The auxiliary storage device 15 is a storage device having a large capacity storage area, such as a hard disk drive (HDD) or a solid state drive (SSD). The auxiliary storage device 15 stores various programs (executable files of programs) and data used for executing the programs. In this embodiment, the auxiliary storage device 15 stores a business process analysis program 15A and a deliverable database 101.

[0026] The communication control device 12 is connected to the network 2 via a communication interface 13, and controls communication with the outside.

[0027] The input / output interface 17 is connected to the output device 18 and the input device 19, and controls the input and output of data between the output device 18 and the input device 19. The input / output interface 17 converts data input from the input device 19 into a data format that can be processed by the processor 11, and converts data output from the processor 11 into a data format that can be output by the output device 18 and outputs the data.

[0028] The deliverable database 101 stores various deliverable data created in a business process. The deliverable data in the deliverable database 101 is stored by the work PC 20, for example.

[0029] Here, the deliverable data may be data created and defined in the process of performing a task, and may be, for example, a part of data in a data table in an arbitrary database. In addition, in the case of the development and design stage, the deliverable data is data such as reports, specifications, study documents, technical documents, drawings, etc. More specifically, the deliverable data is data such as functional specifications, experiment reports, and technical documents describing anticipated issues and countermeasures. In the case of calculations of strength, heat load, cost, etc., the deliverable data is a file such as a spreadsheet software or simulation software created in a past project. In addition, the deliverable data also includes data such as 2D drawings, 3D CAD data, part assembly information, and various attribute information (material, part type, welding instructions, etc.) that indicate the shape and structure. In addition, in the production stage, the deliverable data is data such as mold data, manufacturing process charts, assembly process charts, quality plans, inspection records, and cost estimates. In addition to the above, the deliverable data may include data created in the process of a certain task, such as program code and meeting minutes.

[0030] FIG. 3 is a diagram illustrating an example of a deliverable database according to an embodiment.

[0031] The deliverable database 101 stores entries corresponding to each deliverable data. An entry in the deliverable database 101 includes fields for No., file name, and creation date and time. The No. field stores the number of the deliverable data corresponding to the entry. The file name field stores the file name of the deliverable data corresponding to the entry. The creation date and time field stores the date and time when the deliverable data corresponding to the entry was created.

[0032] Next, the functional configuration of the business process analysis system 1 will be described.

[0033] FIG. 2 is a functional configuration diagram of a business process analysis system according to an embodiment.

[0034] The business process analysis device 10 includes a deliverable database 101, a similar data clustering unit 102, a creation date and time acquisition unit 103, an event log data generation unit 104, a business process estimation unit 106, a task-related application log extraction unit 108, a task separation unit 109, a task activity time calculation unit 110, a task activity time statistics calculation unit 111, and a business process display unit 112. Each functional unit (102 to 112) is configured by the processor 11 reading out a business process analysis program 15A stored in the auxiliary storage device 15 into the main storage device 14 and executing it.

[0035] The worker's PC 20 includes an application log acquisition unit 21 .

[0036] The similar data clustering unit 102 calculates the similarity between each deliverable data stored in the deliverable database 101, and performs clustering (grouping) to classify similar deliverable data (similar data) into the same cluster (group). Specifically, the similar data clustering unit 102 vectorizes each deliverable data, calculates the similarity between the vectors, clusters the deliverable data based on the similarity, and assigns the same type name to the deliverable data in the same cluster. This type corresponds to the task that constitutes the business process. The type name may be a representative file name of the deliverable data in the cluster, or an association table showing the correspondence between the file name or content (actual content of the deliverable data) and the type name may be prepared and used to determine the type name corresponding to the file name. In the latter case, for example, if the content contains the characters "device configuration", the type name may be defined as "creation of device configuration", and such a correspondence may be defined in the association table. Note that this type name is used as a task name in the business process estimation unit 106, so it is preferable to use a name that can be recognized as a task.

[0037] There are various known methods for vectorizing each deliverable data and calculating the similarity, and any of these methods may be used in this embodiment. The point is that it is sufficient to simply calculate the similarity between deliverable data. Deliverable data includes, for example, text documents, image data, spreadsheet data, 3D shape data, and combinations of these.

[0038] For example, an example of a vectorization and similarity calculation method when the deliverable data is a text document is shown below. The text document is divided into words using morphological analysis. Note that this morphological analysis is not necessary when the text document is an English document in which words are separated by spaces. Next, the number of times each word appears in the text document is counted and expressed as a vector. Next, the cosine similarity is calculated for the vectors created for the two text documents. Note that in this example, the similarity is calculated using the frequency of word appearance, but another method for evaluating the importance of words called TF-IDF may also be used.

[0039] Next, we will show an example of a similarity calculation method when the deliverable data is 3D shape data. The shape of the target for which similarity is to be calculated in the 3D shape data is expressed as a voxel structure, several adjacent voxels are defined as a block, and the spectrum is calculated for each block and expressed as a vector. Next, the cosine similarity is calculated for the vectors created for the two 3D shape data.

[0040] In addition, in this embodiment, any clustering method may be used, but for example, the K-means method, the shortest distance method, the multidimensional scaling method, the Ward method, or the like may be used.

[0041] The creation date and time acquisition unit 103 acquires the creation date and time of the deliverable data from the deliverable database 101. The creation date and time acquisition unit 103 acquires, for example, the creation date and time of various files stored in the deliverable database 101. The creation date and time acquisition unit 103 acquires, for example, the last saved date and time (time stamp) as the creation date and time.

[0042] The event log data generator 104 records the tasks constituting the business process and the creation date and time of the deliverable data for each deliverable data as event log data. For example, the event log data generator 104 generates event log data (see FIG. 4) based on the creation date and time of the deliverable data and the type of the deliverable data.

[0043] Fig. 4 is a diagram showing an example of event log data according to an embodiment of the present invention, which is an example of event log data created based on the deliverable database 101 shown in Fig. 3.

[0044] The event log data stores an entry corresponding to deliverable data. An entry of the event log data includes fields for No., case name, deliverable type name (corresponding to a task name), and creation date and time. A number corresponding to the entry is stored in No.. The case name stores the name of the case to which the deliverable data corresponding to the entry is related. The case name may be obtained by identifying it from, for example, the storage location (folder name, etc.) of the deliverable data. Note that the method of obtaining the case name is not limited to this method. The deliverable type name stores a type name assigned to the deliverable data corresponding to the entry by the similar data clustering unit 102. This type name corresponds to the task name. The creation date and time of the deliverable data corresponding to the entry is stored in the creation date and time. In the event log data, in entries of multiple deliverable data classified into the same cluster, the same type name corresponding to the cluster is stored.

[0045] The business process estimation unit 106 generates business process data 105 that expresses task transitions based on the event log data. There are various known methods for generating business process data from event log data, and any of these methods may be used in this embodiment. A representative example of a method for generating business process data from event log data is described below.

[0046] [First step] For each case, the tasks are sorted by the order in which they were performed, i.e., by the order of the creation date and time. For example, the order of tasks for cases 1 to 4 is obtained as shown below. Here, the letters A to E indicate the task names, and (A, B, C) indicate that the tasks were performed in the order A → B → C. Case 1: (A,B,C,D) Case 2: (A,C,D) Case 3: (A,E,C,D) Case 4: (A,B,C,E,C,D)

[0047] [Second step] Extract the task transitions of all cases. For example, the task transitions for case 1 can be expressed as A → B → C → D, as shown in FIG. 5(A). Furthermore, when considering the task transitions for case 2, an order of transition from A to C is added, so the task transitions can be expressed as shown in FIG. 5(B). Furthermore, when considering the task transitions for case 3, an order of transition from A to E and an order of transition from E to C are added, so the task transitions can be expressed as shown in FIG. 5(C). Furthermore, when considering the task transitions for case 4, an order of transition from C to E and an order of transition from E to C are added, so the task transitions can be expressed as shown in FIG. 5(D).

[0048] In this way, the business process estimation unit 106 registers the model expressing the task transition as business process data 105. As models expressing task transition, BPMN (Business Process Model and Notification), Petri nets, UML (Unified Modeling Language) activity diagrams, etc. are known, and any appropriate format may be used for registration. Note that in this embodiment, the type of format of the business process data is not limited, and any format may be used.

[0049] Fig. 6 is a diagram illustrating an example of business process data according to an embodiment of the present invention, which is generated based on the event log data illustrated in Fig. 4 and taking into consideration task transitions.

[0050] According to the business process data, it can be seen that this business process is configured so that a transition occurs from the task “Determine operating conditions” to the task “Create equipment configuration,” from the task “Create equipment configuration” to the task “Create a new design” or the task “Conduct design review,” from the task “Create a new design” to the task “Evaluate performance reliability,” from the task “Evaluate performance reliability” to the task “Evaluate thermal resistance,” and from the task “Evaluate thermal resistance” to the task “Create a new design” or the task “Conduct design review.”

[0051] The application log acquisition unit 107 acquires, as application logs, the dates and times of transition of active windows accompanying the start, end, and switching of applications in the work PC 20, and stores these application logs as files (application log files) in the work PC 20.

[0052] FIG. 7 is a diagram illustrating an example of an application log file according to an embodiment.

[0053] The application log file includes records related to the state of the active window of an application. The application log record includes fields for No., application name, file name, start date and time, end date and time, and PC. No. stores a number indicating the record of the log. The application name stores the name of the application displaying the active window (application name: application name). The file name stores the name of the file displayed in the active window (file name). The application name and file name corresponding to the active window can be extracted from the title of the active window. The start date and time stores the date and time when the window corresponding to the record became active (start date and time). The end date and time stores the date and time when another window became active (end date and time). Here, the time between the start date and time and the end date and time represents the time when the window was active (active time). The PC stores the name (identifier) ​​of the user of the work PC 20 that displayed the corresponding window. For example, record No. 1 indicates that the file Product Specifications.docx was active on Tanaka's work PC in the document editing application from 2022.06.10 10:10 to 2022.06.10 10:20.

[0054] The task-related application log extraction unit 108 performs a task activity analysis process that analyzes the activity and activity time for the deliverable data in the task based on the application log and the creation date and time of the deliverable data. Specifically, the task-related application log extraction unit 108 identifies the time when the deliverable data corresponding to the task was active based on the application log, and analyzes the time before the creation date and time of the deliverable data as a creation activity for the deliverable data (i.e., the corresponding task) while analyzing the time after the creation date and time of the deliverable data as a reference activity for the deliverable data. Note that, when a file that exists in the application log is not stored as deliverable data in the deliverable database 101, the task-related application log extraction unit 108 recognizes the file (non-deliverable data) as data used for reference, and recognizes the activity for the data as a reference activity.

[0055] FIG. 8 is a PAD diagram illustrating a task activity analysis process according to one embodiment.

[0056] First, the task-related application log extraction unit 108 reads an application log file storing an application log from each work PC 20 (S101). Next, the task-related application log extraction unit 108 reads a list of a deliverable data group (each entry) from the deliverable database 101 (S102). Next, the task-related application log extraction unit 108 repeats the processes of steps S104 to S110 for the number of records of the acquired application log, that is, for the records of each acquired application log (S103).

[0057] In the process of steps S104 to S110, first, the task-related application log extraction unit 108 acquires corresponding deliverable data based on the file name of the record of the application log (S104). Next, the task-related application log extraction unit 108 determines whether or not the corresponding deliverable data exists (step S105). As a result, if there is no deliverable file corresponding to the file name of the record (S105: No), it means that the file corresponding to the file name of the record is not deliverable data, so the task-related application log extraction unit 108 regards this file as a file used for reference (non-deliverable data) and recognizes the activity for this file as a reference activity (S106).

[0058] On the other hand, if deliverable data corresponding to the file name of the record in the application log exists (S105: Yes), the task-related application log extraction unit 108 acquires the creation date and time of the existing deliverable data (target deliverable data) (S107). Next, the task-related application log extraction unit 108 determines whether the active time in the record is earlier than the creation date and time of the target deliverable data (S108).

[0059] As a result, for a time period before the creation date and time during the active time (S108: Yes), the task-related application log extraction unit 108 recognizes the activity for the target deliverable data (the task handling the data) during that time as a creation activity (S109). On the other hand, for a time period after the creation date and time during the active time (S108: No), the task-related application log extraction unit 108 recognizes the activity for the target deliverable data (the task handling the data) during that time as a reference activity (S110).

[0060] For example, in the task activity analysis process, when the record No. 2 in the application log shown in FIG. 7 is processed as the processing target, the file "operating condition definition document_Tanaka_220110.docx" was active from "2022.06.10 10:20" to "2022.06.10 12:00". A file corresponding to this file "operating condition definition document_Tanaka_220110.docx" is registered in the deliverable database 101 shown in FIG. 3, that is, it exists as deliverable data and is saved at "2022.06.10 12:00". For this reason, the activity for the file "operating condition definition document_Tanaka_220110.docx" from "2022.06.10 10:20" to "2022.06.10 12:00", which is the active time before the saving date and time, is recognized as a creation activity.

[0061] On the other hand, when processing record No. 4 in the application log, the file "Operation Condition Definition Document_Tanaka_220110.docx" was active from "2022.06.11 09:10" to "2022.06.11 09:40". This file "Operation Condition Definition Document_Tanaka_220110.docx" was saved at "2022.06.10 12:00". Therefore, the activity for the file "Operation Condition Definition Document_Tanaka_220110.docx" from "2022.06.11 09:10" to "2022.06.11 09:40", which is the active time after the save date and time, is recognized as a reference activity.

[0062] Also, for example, when processing record No. 1 or record No. 14 in the application log file, the file of record No. 1, “Product Specifications.docx,” and the file of record No. 14, “CAE Know-How Collection,” are not deliverable data managed in the deliverable database 101, so these files are reference data, and activities during the active time are recognized as reference activities.

[0063] When deliverable data clustered into the same cluster as being similar satisfies a predetermined condition, the task separation unit 109 performs a task separation process for separating a task corresponding to this deliverable data into a task separate from other deliverable data in the same cluster.

[0064] FIG. 9 is a PAD diagram illustrating a task separation process according to one embodiment.

[0065] First, the task separation unit 109 acquires from the event log data a plurality of (for example, two) deliverable data items having the same case name and the same task (S201). Next, the task separation unit 109 judges whether the similarity of the deliverable data items is equal to or lower than a threshold (S202). The threshold here is a value higher than the threshold used for the judgment when clustering the deliverable data items into the same cluster. As a result, if the similarity is equal to or lower than the threshold (S202: Yes), the task separation unit 109 acquires the work PC 20 whose activity for the deliverable data items is the creation activity from the application log (S203). Next, the task separation unit 109 judges whether the user roles of the deliverable data items are different (S204). Here, the user role can be grasped from a table or the like that manages the relationship between the user name and the user role. As a result, if the user roles are different (S204: Yes), the task separation unit 109 acquires the task name for the deliverable data items based on the association table (S205). Next, the task separation unit 109 separates the tasks of these deliverable data according to the task names (S206). Note that if the user roles are not different in S204, nothing is done, that is, the tasks are not separated.

[0066] For example, the entry No. 5 and the entry No. 14 in the event log data shown in Fig. 4 have the same case name, △△ Company, and are the same task of "select device configuration", so in step S201 of the task separation process, these deliverable data are acquired. Here, as shown in the deliverable database 101 shown in Fig. 3, the deliverable data corresponding to No. 5 is "△△ Company\220118_Device Configuration Table_Suzuki.docx", and the deliverable data corresponding to No. 14 is "△△ Company\220118_Device Configuration Table_Yamamoto.docx".

[0067] Also, according to the application log file shown in FIG. 7, it can be seen that the deliverable data "△△ company\220118_equipment configuration table_Suzuki.docx" corresponds to record No. 13 and is active in Suzuki's business PC 20, and the deliverable data "△△ company\220118_equipment configuration table_Yamamoto.docx" corresponds to record No. 37 and is active in Yamamoto's business PC 20. In other words, these deliverable data are used by different business PCs 20 (or users). Here, if the similarity between "△△ company\220118_equipment configuration table_Suzuki.docx" and "△△ company\220118_equipment configuration table_Yamamoto.docx" is equal to or less than the threshold, the process proceeds to step S206 and the tasks are separated.

[0068] Next, an example of a method for defining a task name when a task is separated will be described. As described above, the similar data clustering unit 102 prepares an association table between a file name or content and a task name (type name) in advance, so that, for example, when the content contains the characters "device configuration", the task name is set to "creating device configuration". Below, a method for making it possible to reflect a user role in a task name will be described. In this case, a table (role table) in which a role is defined for each user is prepared in advance. For example, a table is defined with the roles of Suzuki as a person in charge and Yamamoto as an evaluator. In addition, an association table is defined in which, when the content of the deliverable data contains the characters "device configuration" and the user role is "person in charge", the task name is set to "creating device configuration", and when the user role is "evaluator", the task name is set to "reviewing device configuration". Then, when a task name is determined, the role table and the association table are used to define the name of the separated task. For example, in the above example, the task name corresponding to the deliverable data “△△ Company\220118_Equipment Configuration Table_Suzuki.docx” can be “Create equipment configuration,” and the task name for the deliverable data “△△ Company\220118_Equipment Configuration Table_Yamamoto.docx” can be “Review equipment configuration,” so they can be appropriately separated into different tasks.

[0069] 10 is a diagram illustrating an example of business process data after task separation processing according to an embodiment of the present invention, which illustrates the result of performing task separation on the business process data illustrated in FIG.

[0070] With task separation processing, the single task "Create device configuration" in the business process data shown in Figure 6 can be separated into two tasks, "Create device configuration" and "Review device configuration," as shown in Figure 10. This makes it possible to express the business process in more detailed tasks.

[0071] The task separation unit 109 uses the similarity being equal to or less than a predetermined threshold as one of the conditions for determining separation, but this threshold may be received from the user. For example, in the case of document data, the similarity is document similarity, but in the case of document data, it may be necessary to determine that it is not necessary to separate it as a task in the case of minor differences such as typos or corrections to particles. Therefore, the document similarity may be calculated in advance to determine whether the differences are minor, and an input of a threshold that can exclude such minor differences may be received.

[0072] The task activity time calculation unit 110 counts the difference between the start time and the end time of the creation activity for the deliverable data as the creation activity time, and the difference between the start time and the end time of the reference activity as the reference activity time. The task activity time calculation unit 110 creates data that associates the creation activity time with the task corresponding to the deliverable data for which the creation activity was performed, and creates data that associates the reference activity time with the link to the reference source task and the referenced task. Here, for example, the task activity time calculation unit 110 extracts from the application log a creation activity that is performed by the same work PC (user) and immediately after the reference activity, and creates data in which the reference activity time is assigned from the task corresponding to this creation activity (referenced task) to the link to the task corresponding to the reference activity (reference source task).

[0073] Here, a specific example of the processing of the task activity time calculation unit 110 will be described. For example, the activity for the file "operating condition definition_Tanaka_220110.docx" of the record No. 4 in the application log file shown in FIG. 7 is a reference activity, and the immediately following creation activity is a creation activity for "equipment configuration table_Tanaka_220112.docx" of the record No. 5. The task corresponding to "operating condition definition_Tanaka_220110.docx" is "define operating conditions," and the task corresponding to "equipment configuration table_Tanaka_220112.docx" is "create equipment configuration table." Therefore, the task activity time calculation unit 110 associates 0:30:00, which is the activity time of the record No. 4, as the reference activity time for the link from "define operating conditions" (referenced task) to "create equipment configuration table" (reference source task).

[0074] The task activity time statistics calculation unit 111 performs a predetermined statistical process on the creation activity time of each task and the reference activity time of the link between tasks to calculate statistics based on the values ​​of multiple cases (for example, the median, average, maximum, minimum, and variation of each time of multiple cases). Note that the variation may be an index that indicates the variation such as the standard deviation, variance, and range.

[0075] Here, a specific example of the processing of the task activity time statistics calculation unit 111 will be described. For example, the creation activity time of the task "calculate driving conditions" is "1:40:00" for the case "XX Company" (see record No. 2 in the application log shown in FIG. 7), "2:00:00" for "△△ Company" (see record No. 6 in the application log shown in FIG. 7), and "1:00:00" for "□□ Company" (see record No. 26 in the application log shown in FIG. 7). In this case, the task activity time statistics calculation unit 111 calculates the median as 1:40:00, the average as 1:33:20, the maximum as 2:00:00, the minimum as 1:00:00, and the variance as the variation as 0:00:26.

[0076] The business process display unit 112 displays on the output device 18 a business process analysis result screen 200 (see FIG. 11 ) including the task configuration of the business process, time information on the creation activity time of each task (creation activity time or statistics of creation activity time), and time information on the reference activity time between tasks or for specified data (reference activity time or statistics of reference activity time).

[0077] Next, the business process analysis result screen 200 will be described.

[0078] FIG. 11 is a diagram showing an example of a business process analysis result screen according to an embodiment.

[0079] In the business process analysis result screen 200, task components (201, etc.) that indicate each task constituting a business process are arranged, connected by arrows (solid arrows) that indicate the transition relationships of tasks in the business process (the diagram in which these are arranged corresponds to a business process diagram), and reference data components (202, etc.) that indicate data (documents, etc.: data outside the deliverable) that is referenced by the process and different from the deliverable data are arranged, connected by arrows (dashed arrows) that indicate reference relationships between the reference data components and the task components that indicate tasks that reference the reference data components. Also, in the business process analysis result screen 200, task components that indicate tasks in a reference relationship are connected by arrows (dashed arrows) that indicate reference relationships.

[0080] Furthermore, at a predetermined position for the task component (at the lower right position of the task component in FIG. 11), the work activity time or a statistic of the work activity time (for example, the median of the reference activity time) for the task corresponding to the task component is displayed.

[0081] In addition, the arrow indicating the reference relationship displays the reference activity time or a statistic of the reference activity time (median of the reference activity time) in the corresponding reference relationship.

[0082] According to the business process analysis result screen 200, the transition of tasks in a business process can be grasped, and the creation activity time in a task and the reference activity time for referencing reference data or other tasks can be easily grasped.

[0083] Next, the overall processing by the business process analysis device 10 will be described.

[0084] FIG. 12 is a flowchart of the overall processing of a business process analysis device according to an embodiment.

[0085] Before the overall process is executed, a set of deliverable data created by a task is registered in the deliverable database 101 .

[0086] When an instruction to execute the entire process is given, the similar data clustering unit 102 calculates the mutual similarity between each piece of deliverable data and clusters the deliverable data (step S301).

[0087] Next, the creation date and time acquisition unit 103 acquires the creation date and time of the deliverable data from the deliverable database 101 (S302). Next, the event log data generation unit 104 records the tasks constituting the business process and the creation date and time of the deliverable data as event log data for each deliverable data (S303). Next, the business process estimation unit 106 generates business process data indicating the tasks constituting the business process and the transitions of the tasks based on the event log data (S304).

[0088] Next, the business process analysis device 10 performs the process of loop 1 (S305 to S310) for each record of the application log.

[0089] In the process of loop 1, if the file name corresponding to the record of the application log is deliverable data, the task-related application log extraction unit 108 extracts the time during which the deliverable data was active from the record (active time) (S305). The task-related application log extraction unit 108 determines whether or not there is a time before the creation date and time of the deliverable data in the active time (S306). As a result, if there is a time before the creation date and time of the deliverable data in the active time (S306: YES), the task-related application log extraction unit 108 classifies the activity of the time before the creation date and time in the active time as a creation activity for the deliverable data (corresponding task) (S307). Next, the task activity time calculation unit 110 counts up the time classified as a creation activity as a creation activity time for the corresponding task (S308).

[0090] On the other hand, if the active time includes a time later than the creation date and time of the deliverable data (S306: NO), the task-related application log extraction unit 108 classifies the activity in the active time after the creation date and time as a reference activity for the deliverable data (S309). Next, the task activity time calculation unit 110 counts up the time classified as a reference activity as a reference activity time for the deliverable data (between tasks) (S310). Note that if the active time includes a time before and after the creation date and time, both the processes of steps S307 and S308 and steps S309 and S310 are performed.

[0091] After loop 1 processing is executed for all records in the application log, the task activity time statistics calculation unit 111 calculates the creation activity time statistics for each task based on the creation activity time for the task of each case, and calculates the reference activity time statistics between each task based on the reference activity time between the tasks of each case (S311). Next, the business process display unit 112 creates a business process analysis result screen 200 based on the business process data and the calculated creation activity time statistics and reference activity time statistics, and displays the business process analysis result screen on the output device 18 (S312).

[0092] As described above, in this embodiment, the creation activity time of the tasks constituting the business process can be appropriately estimated by the above-mentioned processing.

[0093] The present invention is not limited to the above-described embodiment, and can be modified as appropriate without departing from the spirit of the present invention.

[0094] For example, in the above embodiment, the deliverable database is stored in the business process analysis device 10, but this embodiment is not limited to this. For example, the deliverable database may be stored in a file server or a business PC 20 different from the business process analysis device 10, and the data required for processing may be obtained from the file server or the business PC.

[0095] In addition, in the above embodiment, the application log acquisition unit 107 in each business PC 20 sequentially acquires application logs, and then the business process analysis device 10 acquires the application logs from each business PC 20. However, the present invention is not limited to this. For example, the business process analysis device 10 may sequentially monitor the status of applications in multiple business PCs and acquire the application logs of the multiple business PCs.

[0096] Furthermore, application log acquiring unit 107 may be configured to limit the types of applications recorded as application logs to a predetermined type, such as types other than email applications and chat applications.

[0097] Furthermore, the application log acquisition unit 107 may be configured to record in the application log a non-response time during which no processing is being performed on an application in the work PC 20. Examples of the non-response time include a time during which the work PC 20 is not operated, such as a time spent attending a face-to-face conference or a break. In this case, the task activity time calculation unit 110 may correct the creation activity time and the reference activity time based on the non-response time of the work PC 20. For example, the task activity time calculation unit 110 may correct the creation activity time and the reference activity time by subtracting the non-response time from the creation activity time and the reference activity time.

[0098] Furthermore, in the above embodiment, the task separation unit 109 is provided and task separation processing is performed, but the present invention is not limited to this, and the task separation unit 109 may not be provided and task separation processing may not be performed.

[0099] In the above-described embodiments, a part or all of the processing performed by the processor may be performed by a hardware circuit. The programs in the above-described embodiments may be installed from a program source. The program source may be a program distribution server or a recording medium (e.g., a portable recording medium). [Explanation of symbols]

[0100] 1: business process analysis system, 2: network, 10: business process analysis device, 11: processor, 12: communication control device, 13: communication interface, 14: main memory device, 15: auxiliary memory device, 15A: business process analysis program, 16: internal bus, 17: input / output interface, 18 output device, 19: input device, 20 (20-1, 20-2, . . . 20-n): business PC, 21: application log acquisition unit, 101: deliverable database, 102: similar data clustering unit, 103: creation date and time acquisition unit, 104: event log data generation unit, 106: business process estimation unit, 108: task-related application log extraction unit, 109: task separation unit, 110: task activity time calculation unit, 111: task activity time statistics calculation unit, 112: business process display unit

Claims

1. A business process analysis device that analyzes a business process based on deliverable data created by an application, A processor is provided. The processor, Acquire a creation date and time when the deliverable data was created; obtaining an application log including an active time during which the deliverable data is active in an application; A creation activity time, which is a time during which an activity for creating a deliverable in a task of the business process that creates the deliverable data, is performed, is estimated based on the creation date and time of the deliverable data and the active time. Business process analysis device.

2. The processor, The time before the creation date and time during the active time is counted as the creation activity time. The business process analysis apparatus according to claim 1 .

3. The processor, A reference activity time, which is a time when an activity that refers to the deliverable data was performed, is estimated based on the creation date and time of the deliverable data and the active time. The business process analysis apparatus according to claim 1 .

4. The processor, The time after the creation date and time within the active time is counted as the reference active time. The business process analysis apparatus according to claim 3 .

5. The processor, The time information regarding the creation activity time is displayed in association with the part indicating the task. The business process analysis apparatus according to claim 1 .

6. The processor, Collect multiple deliverable data for multiple cases of the same or similar business processes, Group multiple deliverables into tasks that make up the corresponding business process, generating business process data indicating a transition of the tasks of the business process based on a plurality of deliverable data; displaying a business process diagram including components representing tasks constituting the business process and components representing transitions between tasks according to the business process data; The time information regarding the creation activity time is displayed in association with the part indicating the task of the business process diagram. The business process analysis apparatus according to claim 5 .

7. The application log further includes an active time during which non-deliverable data that is not the deliverable data is active in the application; The processor, estimating the active time for non-deliverable data included in the application log as a reference activity time for the non-deliverable data; The reference activity time of the same non-deliverable data is aggregated; The reference activity time for the non-deliverable data is displayed in association with the part indicating the non-deliverable data. The business process analysis apparatus according to claim 6.

8. The processor, Collect multiple deliverable data for multiple cases of the same or similar business processes, Estimate the production activity time of the deliverable data created in the same task for each project, performing a predetermined statistical process on the estimated production activity time for each item to calculate statistics regarding the production activity time; The calculated statistics regarding the production activity time are displayed as time information regarding the production activity time in association with the part indicating the task. The business process analysis apparatus according to claim 5 .

9. The processor, Collect multiple deliverable data for multiple cases of the same or similar business processes, grouping the plurality of deliverable data into tasks constituting a corresponding business process based on the similarity of the plurality of deliverable data; If the similarity of multiple deliverable data items divided into the same task is lower than a predetermined threshold value and the data items are created by different users, the deliverable data items are separated into multiple different tasks. The business process analysis apparatus according to claim 6.

10. A business process analysis method using a business process analysis device that analyzes a business process based on deliverable data created by an application, comprising: Acquire a creation date and time when the deliverable data was created; obtaining an application log including an active time during which the deliverable data is active in an application; A creation activity time, which is a time during which an activity for creating a deliverable in a task of the business process that creates the deliverable data, is performed, is estimated based on the creation date and time of the deliverable data and the active time. Business process analysis methods.

11. A business process analysis program that causes a computer to execute a process for analyzing a business process based on deliverable data created by an application, The computer includes: obtaining a creation date and time when the deliverable data was created; obtaining an application log including an active time during which the deliverable data is active in the application; A creation activity time, which is a time during which an activity for creating a deliverable in a task of the business process for creating the deliverable data is performed, is estimated based on the creation date and time of the deliverable data and the active time. Business process analysis program.

Images